In around one in three whole-body MRI scans something turns up that nobody was looking for.

In the largest synthesis available the rate was 32.1 percent, with a 95 percent confidence interval of 18.3 to 50.1. The analysis pools 12 studies covering 5,373 people without symptoms. It refers to critical and indeterminate findings together, split into 13.4 percent critical and 13.9 percent indeterminate. What matters for interpretation is that the rate depends heavily on the imaging protocol: studies including vascular or colon imaging reached 49.7 percent, studies without those sequences 23.0 percent. Quoting a single percentage as a binding rate would therefore mislead. The defensible statement is this: in whole-body MRI an incidental finding is an expected outcome, not an exception.

Most are not. A meta-analysis of 32 studies with 27,643 participants found potentially serious incidental findings in 3.9 percent of brain and body MRI examinations, with a very wide confidence interval of 0.4 to 27.1. Including findings of uncertain seriousness, the share rises to 12.8 percent. About half the potentially serious findings were suspected malignancies. The real risk lies less in the finding itself than in the chain it sets off: further imaging, sometimes a tissue sample, occasionally a procedure that turns out in hindsight to have been unnecessary. That is why the question of what a finding would lead to belongs before the examination, not after it.

A finding does not reach you as raw data but with medical interpretation: how likely it is to mean something, which explanations may sit behind it, and what the sensible next step is. An abnormal image or value is a reason for work-up here and never a diagnosis. The work-up itself continues within specialist standard care, which is where we hand over. We do not replace that path and deliberately offer no procedure whose purpose is to keep a finding in-house. Results from procedures we classify as research-adjacent are documented and contextualised, but on their own they do not trigger treatment.

It argues against a check-up sold without this information. The trade-off looks different for everyone. For someone with a family history or a concrete risk profile, broad examination can make sense. For someone without risk indicators and with a low tolerance for uncertainty, the most likely outcome is a work-up chain that ends with nothing. More important than whether you get examined is which parts of an examination rest on solid evidence. Blood pressure, lipids including ApoB, early detection of diabetes, cardiorespiratory fitness and established cancer screening do. Whole-body imaging does not, not with that degree of certainty, and we do not claim otherwise.

Less reliable than most people assume, and this is the most uncomfortable figure on the page. In the Kwee and Kwee meta-analysis, not one of the 12 included studies verified negative findings beyond five years. How often a normal whole-body MRI actually stayed normal over the longer term is therefore simply not studied. Only one study reported false negatives, at a rate of 2.0 percent. In practice: a normal result is a snapshot, not a clean bill of health. In particular it does not replace established screening with its fixed intervals, which is designed around intervals rather than a single examination for exactly this reason.